Article Per Year (5 Year)
p-Index From 2021 - 2026
1.037
P-Index
This Author published in this journals
All Journal ILKOMNIKA: Journal of Computer Science and Applied Informatics Journal of Innovation Research and Knowledge Jurnal ELTEK
Adzikirani, Adzikirani
Unknown Affiliation
Humanities Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Electrical & Electronics Engineering Engineering Health Professions Law, Crime, Criminology & Criminal Justice Social Sciences

Published : 5 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 5 Documents
Search

 1 
Sistem pemesanan pintar di Pujasera menggunakan QR code dan smartphone Rasyid, Abdul; Suharto, Nugroho; Adzikirani, Adzikirani
JURNAL ELTEK Vol. 20 No. 2 (2022): Oktober 2022
Publisher : Politeknik Negeri Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33795/eltek.v20i2.360

Abstract

Pada proses pemesanan makanan di pujasera terkadang terdapat beberapa kesulitan yang menyebabkan terhambatnya proses pengantaran makanan diantaranya adalah sulitnya mengidentifikasi pelanggan yang memesan, dimana letak posisi pemesan, dan lamanya proses yang terjadi. Pada proses pemesanan, makanan menggunakan sistem konvensional, pelayan harus meneriakan nama pemesan untuk mencari tempat pemesan. Pada sistem pemesanan pujasera sendiri terdiri dari beberapa tahapan, mulai dari pemilihan kedai, pemesanan minuman atau makanan sampai makanan telah siap disajikan. Pada penelitian ini dibuat suatu sistem pemesanan pintar pujasera menggunakan QR code dan aplikasi android. Untuk pemesanan pelanggan harus memindai QR code dan kemudian memilih kedai, makanan dan minuman yang akan dipesan. Setelah dibayar, data pemesanan berupa makanan, minuman, nama pelanggan dan nomor meja akan dikirimkan ke masing-masing kedai untuk di proses dan setelah pesanan selesai disiapkan, sistem akan mengirimkan notifikasi ke pelayan agar pesanan segera diantar oleh pelayan ke pemesan. Sistem yang dibuat menggunakan software pendukung yaitu kodular. Untuk mendeteksi pesanan yang sudah disiapkan menggunakan sensor hc-SR04, modul wifi esp8266, dan buzzer. Selanjutnya akan dikirimkan ke database firebase dan akan di tampilkan pada aplikasi web di kasir. ABSTRACT In ordering food at food courts, sometimes some difficulties cause too many delays in the food delivery process. Among them is the problem of identifying the customer who collected, the position of the orderer, and the length of the process. For example, in ordering food using a conventional system, the waiter must shout the customer's name to find a place to call. The food court ordering system consists of several stages, from choosing a shop to ordering drinks or food until the food is ready to be served. This study created an intelligent food court ordering system using a QR code and an android application. For customer orders, they must scan the QR code and select the shop, food and drink to be ordered. After being paid, the collected data in the form of food, drinks, customer names and table numbers will be sent to each shop to be processed and after the order has been prepared, the system will send a notification to the waiter so that the waiter immediately delivers the order to the customer. The system is made using supporting software, namely modular. To detect orders that have been prepared using the HC-SR04 sensor, esp8266 wifi module, and buzzer. The order and notification will then be sent to the firebase database and displayed on the web application at the checkout.
Optimasi penerjemahan bahasa asing dengan teknologi IoT pada kelas internasional Politeknik Negeri Malang Ardiansyah, Rizky; Bastiar, Berliana; Adzikirani, Adzikirani; Marya, Dianthy; Novianti, Atik
JURNAL ELTEK Vol. 23 No. 1 (2025): April 2025
Publisher : Politeknik Negeri Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33795/eltek.v23i1.6951

Abstract

Perbedaan bahasa mencerminkan keberagaman budaya dan sejarah suatu negara. Memahami perbedaan ini memperkaya wawasan dan meningkatkan komunikasi global. Dalam rapat internasional, peserta yang tidak memahami bahasa asing sering mengalami kesulitan dalam berpartisipasi aktif, terutama di ruang tertutup, sehingga menghambat pencapaian tujuan diskusi. Penelitian ini mengembangkan sistem Speech to Text berbasis deep learning, menggunakan Mel-Frequency Cepstral Coefficients (MFCC) untuk menganalisis sinyal suara dan Convolutional Neural Network (CNN) untuk meningkatkan akurasi pengenalan serta koreksi ejaan. Raspberry Pi digunakan sebagai penghubung antara headset dan sistem, memungkinkan penerjemahan real-time. Sistem ini juga mengintegrasikan DeepL API sebagai mesin penerjemah dan Text to Speech untuk menghasilkan keluaran suara. Sebagai studi kasus, pengujian dilakukan di Politeknik Negeri Malang kelas internasional, dengan peserta yang terdiri dari mahasiswa dan pengajar yang memiliki latar belakang bahasa yang berbeda. Hasil pengujian menunjukkan metode CNN mencapai rata-rata akurasi 98%, membuktikan efektivitasnya dalam meningkatkan transkripsi dan penerjemahan otomatis. Sistem ini mampu mendukung komunikasi lintas bahasa secara efisien dan real-time, memfasilitasi pemahaman dalam diskusi multibahasa, dan meningkatkan interaksi di lingkungan akademis internasional.   ABSTRACT Language differences reflect the cultural and historical diversity of a country. Understanding these differences enriches knowledge and enhances global communication. In international meetings, participants who do not understand foreign languages often face difficulties in actively participating, especially in closed settings, thus hindering the achievement of discussion goals. This research develops a Speech to Text system based on deep learning, using Mel-Frequency Cepstral Coefficients (MFCC) to analyze sound signals and Convolutional Neural Networks (CNN) to improve accuracy in recognition and spelling correction. A Raspberry Pi is used as a connector between the headset and the system, enabling real-time translation. The system also integrates the DeepL API as the translation engine and Text to Speech for generating audio output. As a case study, testing was conducted at Politeknik Negeri Malang's international class, with participants consisting of students and instructors from diverse language backgrounds. The test results showed that the CNN method achieved an average accuracy of 98%, demonstrating its effectiveness in improving transcription and automatic translation. This system effectively supports real-time, cross-language communication, facilitates understanding in multilingual discussions, and enhances interaction in the international academic environment.
ANALISIS PREDIKTIF MENGGUNAKAN MACHINE LEARNING UNTUK MENANGGULANGI MASALAH REJECT PRODUK PADA PROSES PRODUKSI PT. XYZ Adevian Fairuz Pratama; Dhebys Suryani; Firgi Sotya Izzudin; Adzikirani, Adzikirani
Journal of Innovation Research and Knowledge Vol. 4 No. 5: Oktober 2024
Publisher : Bajang Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.53625/jirk.v4i5.8725

Abstract

PT. XYZ adalah perusahaan makanan dan minuman di Indonesia, terkenal dengan minuman isotoniknya yang menggantikan cairan tubuh saat aktivitas fisik. Produk ini populer di kalangan konsumen lokal dan internasional karena formulanya yang inovatif. PT. XYZ sangat menekankan kualitas, sehingga produk yang tidak memenuhi standar ketat, seperti tekanan botol, akan di-reject untuk memastikan keamanan dan kualitas. Penelitian ini menggunakan metode machine learning, khususnya XGBoost, untuk analisis prediktif. Data diambil dari mesin Liquid Nitrogen dan Bottle Pressure Detector, mencakup pressure bottle, temperatur tangki, sensor liquid nitrogen, dan lainnya pada tahun 2023. Data diolah dan dianalisis untuk memastikan kesesuaian dengan kebutuhan PT. XYZ dan kecocokan dengan metode XGBoost. Business Understanding mengidentifikasi tujuan bisnis dan dampak deteksi tekanan botol. Data Understanding dilakukan dengan Exploratory Data Analysis (EDA). Data dibagi menjadi 80% untuk training dan 20% untuk testing. Data Preparation melibatkan preprocessing untuk membersihkan duplikasi dan missing value. Modeling menggunakan algoritma XGBoost dengan parameter learning_rate sebesar 0.5, gamma sebesar 2, max_depth sebesar 6, n_estimators sebesar 100, colsample_bytree sebesar 0.4, subsample sebesar 0.7, reg_lambda sebesar 3, dan min_child_weight sebesar 1 terbukti terbaik. Evaluasi menunjukkan nilai MAPE 6.88% dan akurasi 93.12%. Prediksi menunjukkan jumlah reject sekitar 23 pada 1 Januari 2024 pukul 07:00, dan sekitar 22 pukul 08:00. Rekomendasi untuk penelitian selanjutnya adalah menggunakan lebih banyak data dan menggabungkan XGBoost dengan model machine learning lainnya
Development of a Driver Drowsiness Detection System Using YOLO, EAR, and MAR for Driving Performance Evaluation Mahfudi, Isa; Rasyid, Muhammad Akida Jabbar; Adzikirani, Adzikirani; Kusumawardani, Mila; Soelistianto, Farida Arinie; Wijayanti, Rieke Adriati; Novianti, Atik
ILKOMNIKA Vol 8 No 2 (2026): Volume 8, Number 2, August 2026
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28926/ilkomnika.v8i2.878

Abstract

Driver fatigue and drowsiness are among the major factors causing traffic accidents, leading to decreased alertness and slower reaction time. This study aims to develop a computer vision-based driver drowsiness detection system by integrating the You Only Look Once (YOLO), Eye Aspect Ratio (EAR), and Mouth Aspect Ratio (MAR) methods, as well as to evaluate driving performance quantitatively. The system is designed to operate in real time using a camera as the visual data source, which is processed through face detection, facial landmark extraction, and the calculation of EAR and MAR parameters as indicators of eye condition and yawning activity. The experimental results show that the system is capable of classifying driver conditions into three categories: normal (alert), drowsy, and microsleep, with an average F1-score of 0.92. In addition, this study proposes the Alertness Index (AI) as a composite indicator calculated based on drowsiness frequency, eye closure duration, and yawning intensity to represent the driver’s alertness level more comprehensively. The system is also able to operate in real time with stable performance, supporting implementation in real-world conditions. The results demonstrate that the integration of YOLO, EAR, and MAR not only improves detection accuracy but also enables objective and data-driven driving performance evaluation. This system can be applied as a driver monitoring solution to improve road safety.
ZCR-Based Suspicious Sound-Event Monitoring for Examination Supervision Using Android Smartphones Adzikirani, Adzikirani; Ardiansyah, Rizky; Rasyid, Abdul; Pratama, Adevian Fairuz
ILKOMNIKA Vol 8 No 2 (2026): Volume 8, Number 2, August 2026
Publisher : Lembaga Penelitian dan Pengabdian Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28926/ilkomnika.v8i2.904

Abstract

This study presents a functional prototype of an Android-based classroom noise-mapping system that uses smartphone microphones, Zero Crossing Rate (ZCR), and Firebase Realtime Database. Two integrated applications were developed: a student-side sensing client and a lecturer monitoring dashboard. Functional testing covered three source conditions (human speech, a dropped pen, and table movement) at three source desks. Each event was observed at seven desk positions, producing 63 sensor-location records from nine source events. The active source position was detected in all nine events. Source-classification accuracy was 77.8% (7/9): human speech and dropped-pen events were correctly classified in all trials, while only one of three table-movement events was classified as an object and two were returned as unclear. The results demonstrate real-time synchronization and functional spatial monitoring, while also showing sensitivity to sound propagation and overlapping acoustic patterns. Because the documented prototype uses an uncalibrated level indicator and its appendix retains a simulated audio-buffer routine, the findings should be interpreted as functional prototype validation rather than calibrated sound-level or field-performance validation.
Co-Authors Abdul Rasyid Adevian Fairuz Pratama ATIK NOVIANTI Bastiar, Berliana Dhebys Suryani Hormansyah, Dhebys Suryani Firgi Sotya Izzudin Isa Mahfudi Kusbianto P. A, Deddy Kusbianto P. A Kusumawardani, Mila Marya, Dianthy Pratama, Adevian Fairuz Rasyid, Muhammad Akida Jabbar Rieke Adriati Wijayanti RIZKY ARDIANSYAH Rosa Andrie Asmara Soelistianto, Farida Arinie Suharto, Nugroho